An apparatus comprising a processor and a non-transitory computer-readable storage medium storing instructions, when executed by the processor, cause the apparatus to perform a method comprising determining a first and second set of coordinates, associated with a first and second feature, respectively, in one or more first and second images, respectively, the first and second images are captured from a first and second position and/or orientation relative to the physical environment, respectively, re-projecting the first set of coordinates to one or more 2D spaces associated with the second images, comparing the re-projected first set of coordinates with the second set of coordinates in at least one of position closeness, feature closeness, or stereo constraints to determine a correspondence between the first and second features and determining a change between the first and second orientations and/or positions with respect to the physical environment based on the determined correspondence.